Why Grokster and Morpheus Won, Why Napster Lost, and What the Future of Peer-to-Peer File Sharing Looks Like Now

On April 25, in M-G-M v. Grokster, U.S. District Judge Stephen Wilson dismissed a copyright infringement lawsuit brought by a group of movie studios and record companies against Grokster and Morpheus. (Grokster and Morpheus are peer-to-peer services that enable users to share copyrighted music, movies, and other content over the Internet without paying a dime to the copyright owners.)

Many observers were surprised. They had assumed that Grokster and Morpheus would - like Napster in A&M Records v. Napster - be shut down for facilitating individual file sharers' copyright infringement. But Judge Wilson, after carefully examining the underlying technology, found that though users' infringement was occurring, Grokster and Morpheus were not contributing to or authorizing it. Thus, they could not be held liable.

The decision is obviously bad news for Hollywood studios and record companies. If it is upheld on appeal, they will continue to face wide-scale infringement of their copyrights.

If the decision is indeed upheld on appeal, will that be good news for consumers? That is a more complicated question. The answer depends heavily on Hollywood's reaction. Will it continue its battle on other fronts - focusing perhaps not on the services, but on their users? Or will it, instead, launch new strategies to take advantage of the powerful business opportunities that peer-to-peer might provide?

Comparing and Contrasting Grokster, Morpheus, and Napster

To see what is likely to occur in the future, it's helpful first to take a closer look at the differences between Grokster, Morpheus, and Napster.

First, Grokster. It offers for download a branded version of software owned by Sharman Networks, a company incorporated in Vanuatu - a remote Pacific island chain that markets itself as protecting corporate secrecy.

When a user boots the software, his computer is directed to sign on to a "root supernode" (a server owned by Sharman), which then directs the user to a "local supernode." The "local supernode" is some user's computer, which has been temporarily designated to route file-sharing requests among a large number of other users. (A particular user's computer may function as a local supernode one day but not the next; the process is largely invisible to the user).

Suppose a Grokster user requests a certain file - it could be a song, a movie clip, a video game, or an e-book. His search request is relayed among a large number of local supernodes and on to individual users. Once the requested file is found, it is transferred directly between the users.

Now let's look at Morpheus. Its software is based on the Gnutella peer-to-peer platform, built from "open source" code. Morpheus users connect to the Gnutella network by contacting another user who is already connected. (This initial connection is usually made by linking to a computer on the network that maintains a constantly changing list of IP addresses for certain currently active nodes.)

The Gnutella network is a "pure" peer-to-peer network - composed of users running Gnutella-compatible software such as LimeWire, BearShare and Shareaza. It does not use supernodes. Instead, user search requests are passed from user to user in the network until the requested file is found. The file is then transferred directly between the two users.

So what's the difference between Grokster and Morpheus, on one hand, and Napster, on the other? It is this: when Grokster and Morpheus users search for and receive digital files, they do so without information being relayed to or by any computer owned or controlled by Grokster or Morpheus. Thus, as the district court noted, if Grokster or Morpheus shut down, their users could continue to share files with little or no disruption.

In contrast, Napster users relayed search requests through a central server owned by Napster (the Napster central server also maintained an index of users and files available on the network at any given time).

Why the Difference In Network Architecture Mattered in Court

This difference in network architecture changed the result in the two court cases. Why?

In both cases, Hollywood sued file sharing networks for contributory and vicarious copyright infringement - alleging that they both aided, and were indirectly responsible for (by failing to control) user piracy. With respect to both claims, the difference was dispositive.

The Contributory Infringement Issue

Napster clearly facilitated its users' infringement. It ran the "site and facilities" over which users exchanged files; it knew of the infringement; and it had at least some power to block access to users who supplied infringing material to others, but did not do so.

In contrast, Grokster and Morpheus do not provide the "site and facilities" for direct infringement; they simply provide the software, and then get out of the way. Some users employ that software to infringe copyrights. But others use it to lawfully exchange uncopyrighted materials such as public domain songs, software, and e-books.

In the district court's view, Grokster and Morpheus are thus similar to companies selling copy machines or VCRs, which also can be used to infringe copyright. Both indirectly enable infringement, but neither should be held liable for contributory infringement. The Court reasoned as follows:

The Vicarious Infringement Issue

For similar reasons, Grokster and Morpheus escaped liability for vicarious infringement. The court held that since they lacked the ability to supervise or control file sharing over the Sharman or Gnutella networks, or to restrict access to those networks, they could not be held indirectly responsible for their users' infringement on the theory that it was, in essence, their own.

The plaintiffs argued that the defendants should have altered their software to prevent users from sharing copyrighted files, and because they didn't, they remained vicariously liable for the sharing. But the district court rejected that contention, holding that "[t]he doctrine of vicarious infringement does not contemplate liability based upon the fact that a product could be made such that it is less susceptible to unlawful use, where no control over the user of the product exists."

What Will Hollywood Do Next? The Option of Going After Users

The district court's careful application of well-established copyright rules makes it reasonably likely that the decision holding that Grokster and Morpheus are not liable will be affirmed on appeal. If so, what will Hollywood do?

One option would be for Hollywood to begin to attack individual peer-to-peer users. Indeed, the major record labels have already taken some significant steps in that direction.

To take one example, the Recording Industry successfully sought from Internet Service Provider Verizon the name of a KaZaA peer-to-peer network user who allegedly has shared hundreds of copyrighted songs. In January, U.S. District Judge John Bates issued a ruling interpreting the Digital Millennium Copyright Act to require Verizon and other ISPs to turn over information about subscribers upon receipt of a subpoena listing an alleged infringer's IP address. (Troublingly, according to the ruling, there is no requirement that a lawsuit be filed or that the subpoena be signed or reviewed by a judge.)

To take another example, the record companies sued four students accused of running file sharing networks on their university campuses. The students settled for between $12,000 and $17,500 each. And there may be more such suits to come: Last week, the recording industry sent hundreds of thousands of messages to file traders on Grokster and KaZaA, warning them that they could be readily identified and face legal prosecution.

Another Option: Cyberwarfare

If Hollywood really wants to get rough, they might get down to the ugly task of cyberwarfare.

Perhaps the most mild tactic in such warfare was recently chosen by Madonna, who has been flooding file-sharing services with dummy files asking "What the f*ck do you think you're doing?" (In response, an unknown hacker posted downloads from Madonna's latest album on Madonna's own website, along with a tartly worded reply to Madonna's inquiry.)

But much more noxious tactics may yet be employed. Peer-to-peer's efficiency at distributing digital files can also be exploited to distribute files containing malicious programs such as Trojan horses and viruses. And a recent article in the New York Times details quiet efforts by the major record companies to develop software programs that would interfere with - and, in some cases, damage - computers engaged in file sharing.

Of course, resorting to such extreme measures would almost certainly inflame public opinion. And, as George Washington University Law School Professor Orin Kerr has argued, it might even land a few record company executives in jail, absent Congressional immunity for such tactics, which (believe it or not) has been proposed.

There must be a better way.

One Promising Option: Apple's Affordable Permanent Downloads

After litigating Napster out of business, the record industry created a number of legal music download services, such as Pressplay, MusicNet and MusicNow. The services, however, are flawed.

They offer limited access to a limited library of songs for a not-so-limited subscription fee of between $10 and $20 per month. And typically, they only allow for music "rental" that lasts only until you stop paying the subscription fees. Seldom do they allow the "burning" of songs to a CD. Not surprisingly, these services have not caught on.

Enter Apple Computer's new iTunes Music Store - a new and improved legal download service. It allows you to buy (not just "rent") music for $.99 per song, or $10.00 per album - significantly cheaper than the cost of a typical CD. Songs bought can be downloaded to a Mac or Apple iPod portable music player, or burned onto a CD.

In its first week in business, Apple sold more than one million songs - despite the fact that the iTunes service is available only to users of Apple computers, which have less than 4 percent share of the U.S. market. Meanwhile, Apple has announced that it is adding additional music to its library, and will introduce a version of its service for Windows machines by year-end.

If Apple's strong numbers continue, they will demonstrate an enormous pent-up demand for reasonably-priced online music. They will also prove that most people do not want to steal music - if given a workable online opportunity, consumers want to pay.

Why Apple's Solution Benefits Both Producer and Consumer

Hollywood should take note of Apple's success, and build on it. After all, online distribution will benefit companies, not just consumers.

It has been estimated that distribution and manufacturing costs account for between a third and half of the cost of a typical CD at retail. Distribution and manufacturing costs likely comprise a similar percentage of the retail price for movies sold on DVD. Online distribution could slash these costs, allowing the price of music and movies to come down significantly. The result will be to expand output and - importantly to Hollywood - profits.

The lower-cost online environment would also spur sales and innovation in net "appliances," such as portable music devices like iPod. It will similarly encourage content packages and formats that offer more flexibility and higher audio fidelity than clunky and relatively crude-sounding CDs.

Hollywood's future is online. The question now is whether Hollywood's setback in Grokster will bring that future closer, or push it further off. Stay tuned.